Artificial neural network controller for maintaining underwater swarm robots' wireless connections

• Main Authors: Vicmudo, Marck P., Dadios, Elmer P.
• Format: text
• Published: Animo Repository 2016
• Subjects: Swarm intelligence; Autonomous underwater vehicles; Wireless communication systems; Mechanical Engineering; Robotics
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/340; https://animorepository.dlsu.edu.ph/context/faculty_research/article/1339/type/native/viewcontent
• Institution: De La Salle University

Underwater swarm robots will enable various applications such as underwater environment monitoring, gathering of data and search and rescue mission. These robots serves as sensor networks comprises of large number of wireless sensor nodes communicating to each other within a coverage range of distance. Thus, maintaining these coverage positions of each robot plays important roles to prolong the network task. Using artificial neural network, the robot will adapt the wireless communication conditions in environment and maintain and establish desirable wireless communication connections for each swarm robots. This will ensure that each robot will be used for data transferring since they maintain wireless connections. Each robot proposed a neural network controller. The quality of the wireless link is measured through the received signal strength indicator. The local neural network is trained for controlling its movement to set up and achieve the good wireless link quality with its neighbors. Thus, these movement effects of all robots will enable the swarm robot system to maintain the preferred communication coverage. © 2015 IEEE.